ARCANE Reweighting: A Monte Carlo Technique to Tackle the Negative Weights Problem in Collider Event Generation

TL;DR

This paper presents ARCANE reweighting, a Monte Carlo method that reduces negative weights in collider event simulations, improving computational efficiency without biasing physical observable distributions.

Contribution

The paper introduces ARCANE reweighting, a novel, exact Monte Carlo technique to mitigate negative weights in collider event generation, enhancing simulation efficiency.

Findings

ARCANE reweighting effectively reduces negative weights in simulations.

The method is exact and unbiased in physical observable distributions.

Demonstrated on a physics example in a companion paper.

Abstract

Negatively weighted events, which appear in the Monte Carlo (MC) simulation of particle collisions, significantly increases the computational resource requirements of current and future collider experiments. This paper introduces and theoretically discusses an MC technique called ARCANE reweighting for reducing or eliminating negatively weighted events. The technique works by redistributing (via an additive reweighting) the contributions of different pathways within an event generator that lead to the same final event. The technique is exact and does not introduce any biases in the distributions of physical observables. A companion paper demonstrates the technique for a physics example.